Dataset Viewer

What is Dataset Viewer?

Dataset Viewer is an MCP (Multi-Cloud Platform) server designed to interact with the Hugging Face Dataset Viewer API. It enables users to browse, analyze, and manage datasets hosted on the Hugging Face Hub, providing a robust interface for data exploration and manipulation.

How to use Dataset Viewer?

To use Dataset Viewer, you first need to install it. This involves cloning the repository, setting up a Python virtual environment with uv, and installing the project in development mode. Once installed, you can configure it by setting the HUGGINGFACE_TOKEN environment variable for private dataset access. For integration with Claude Desktop, you can add a specific configuration to your claude_desktop_config.json file, specifying the command and arguments to run the server.

Key Features of Dataset Viewer

Resources

• Uses dataset:// URI scheme for accessing Hugging Face datasets.
• Supports dataset configurations and splits.
• Provides paginated access to dataset contents.
• Handles authentication for private datasets.
• Supports searching and filtering dataset contents.
• Provides dataset statistics and analysis.

Tools

Dataset Viewer offers a suite of tools for dataset interaction:

• validate: Checks if a dataset exists and is accessible.
• get_info: Retrieves detailed information about a dataset.
• get_rows: Fetches paginated contents of a dataset.
• get_first_rows: Gets the initial rows from a dataset split.
• get_statistics: Provides statistical data about a dataset split.
• search_dataset: Allows text searching within a dataset.
• filter: Filters rows using SQL-like conditions.
• get_parquet: Downloads the entire dataset in Parquet format.

Use Cases of Dataset Viewer

Dataset Viewer is ideal for:

• Data Scientists and Researchers: Quickly exploring and understanding the structure and content of Hugging Face datasets.
• Machine Learning Engineers: Accessing and preparing datasets for model training and evaluation.
• Data Analysts: Performing ad-hoc queries and statistical analysis on large datasets without needing to download them entirely.
• Developers: Building applications that require programmatic access and manipulation of Hugging Face datasets.

FAQ from Dataset Viewer

Q: What are the prerequisites for installing Dataset Viewer? A: You need Python 3.12 or higher and uv (a fast Python package installer and resolver).

Q: How do I access private Hugging Face datasets? A: You need to set the HUGGINGFACE_TOKEN environment variable with your Hugging Face API token.

Q: Can I filter dataset rows using SQL-like queries? A: Yes, the filter tool allows you to use SQL WHERE and ORDER BY clauses to filter dataset rows.

Q: How can I integrate Dataset Viewer with Claude Desktop? A: You can add a specific mcpServers configuration to your Claude Desktop config file, pointing to the Dataset Viewer server command.

Dataset Viewer MCP Server

An MCP server for interacting with the Hugging Face Dataset Viewer API, providing capabilities to browse and analyze datasets hosted on the Hugging Face Hub.

Features

Resources

• Uses dataset:// URI scheme for accessing Hugging Face datasets
• Supports dataset configurations and splits
• Provides paginated access to dataset contents
• Handles authentication for private datasets
• Supports searching and filtering dataset contents
• Provides dataset statistics and analysis

Tools

The server provides the following tools:

1. validate

   • Check if a dataset exists and is accessible
   • Parameters:
     • dataset: Dataset identifier (e.g. 'stanfordnlp/imdb')
     • auth_token (optional): For private datasets

2. get_info

   • Get detailed information about a dataset
   • Parameters:
     • dataset: Dataset identifier
     • auth_token (optional): For private datasets

3. get_rows

   • Get paginated contents of a dataset
   • Parameters:
     • dataset: Dataset identifier
     • config: Configuration name
     • split: Split name
     • page (optional): Page number (0-based)
     • auth_token (optional): For private datasets

4. get_first_rows

   • Get first rows from a dataset split
   • Parameters:
     • dataset: Dataset identifier
     • config: Configuration name
     • split: Split name
     • auth_token (optional): For private datasets

5. get_statistics

   • Get statistics about a dataset split
   • Parameters:
     • dataset: Dataset identifier
     • config: Configuration name
     • split: Split name
     • auth_token (optional): For private datasets

6. search_dataset

   • Search for text within a dataset
   • Parameters:
     • dataset: Dataset identifier
     • config: Configuration name
     • split: Split name
     • query: Text to search for
     • auth_token (optional): For private datasets

7. filter

   • Filter rows using SQL-like conditions
   • Parameters:
     • dataset: Dataset identifier
     • config: Configuration name
     • split: Split name
     • where: SQL WHERE clause (e.g. "score > 0.5")
     • orderby (optional): SQL ORDER BY clause
     • page (optional): Page number (0-based)
     • auth_token (optional): For private datasets

8. get_parquet

   • Download entire dataset in Parquet format
   • Parameters:
     • dataset: Dataset identifier
     • auth_token (optional): For private datasets

Installation

Prerequisites

• Python 3.12 or higher
• uv - Fast Python package installer and resolver

Setup

1. Clone the repository:

git clone https://github.com/privetin/dataset-viewer.git
cd dataset-viewer

2. Create a virtual environment and install:

# Create virtual environment
uv venv

# Activate virtual environment
# On Unix:
source .venv/bin/activate
# On Windows:
.venv\Scripts\activate

# Install in development mode
uv add -e .

Configuration

Environment Variables

• HUGGINGFACE_TOKEN: Your Hugging Face API token for accessing private datasets

Claude Desktop Integration

Add the following to your Claude Desktop config file:

On Windows: %APPDATA%\Claude\claude_desktop_config.json

On MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "dataset-viewer": {
      "command": "uv",
      "args": [
        "--directory",
        "parent_to_repo/dataset-viewer",
        "run",
        "dataset-viewer"
      ]
    }
  }
}

License

MIT License - see LICENSE for details